In cereals, starch makes up approximately 45-65% of the weight of the mature grain. The starch is composed of two types of molecule, amylose and amylopectin. Amylose is an essentially linear molecule composed of α-1,4 linked glucosidic chains, while amylopectin is highly branched with α-1,6 glucosidic bonds linking linear chains.
The synthesis of starch in the endosperm of higher plants is carried out by a suite of enzymes that catalyse four key steps. Firstly, ADP-glucose pyrophosphorylase activates the monomer precursor of starch through the synthesis of ADP-glucose from G-1-P and ATP. Secondly, the activated glucosyl donor, ADP-glucose, is transferred to the non-reducing end of a pre-existing α1-4 linkage by starch synthases. Thirdly, starch branching enzymes introduce branch points through the cleavage of a region of α-1,4 linked glucan followed by transfer of the cleaved chain to an acceptor chain, forming a new α-1,6 linkage. Starch branching enzymes are the only enzymes that can introduce the α-1,6 linkages into α-polyglucans and therefore play an essential role in the formation of amylopectin. Finally, starch debranching enzymes remove some of the branch linkages although the mechanism through which they act is unresolved (Myers et al., 2000).
While it is clear that at least these four activities are required for normal starch granule synthesis in higher plants, multiple isoforms of each of the four activities are found in the endosperm of higher plants and specific roles have been proposed for individual isoforms on the basis of mutational analysis (Wang et al, 1998, Buleon et al., 1998) or through the modification of gene expression levels using transgenic approaches (Abel et al., 1996, Jobling et al., 1999, Sewall et al., 2000). However, the precise contributions of each isoform of each activity to starch biosynthesis are still not known, and it is not known whether these contributions differ markedly between species. In the cereal endosperm, two isoforms of ADP-glucose pyrophosphorylase are present, one form within the amyloplast, and one form in the cytoplasm (Denyer et al., 1996, Thorbjornsen et al., 1996). Each form is composed of two subunit types. The shrunken (sh2) and brittle (bt2) mutants in maize represent lesions in large and small subunits respectively (Giroux and Hannah, 1994). Four classes of starch synthase are found in the cereal endosperm, an isoform exclusively localised within the starch granule, granule-bound starch synthase (GBSS), two forms that are partitioned between the granule and the soluble fraction (SSI, Li et al., 1999a, SSII, Li et al., 1999b) and a fourth form that is entirely located in the soluble fraction, SSIII (Cao et al, 2000, Li et al., 1999b, Li et al., 2000). GBSS has been shown to be essential for amylose synthesis (Shure et al., 1983), and mutations in SSII and SSIII have been shown to alter amylopectin structure (Gao et al, 1998, Craig et al., 1998). No mutations defining a role for SSI activity have been described.
Three forms of branching enzyme are expressed in the cereal endosperm, branching enzyme I (SBEI), branching enzyme IIa (SBEIIa) and branching enzyme IIb (SBEIIb) (Hedman and Boyer, 1982, Boyer and Preiss, 1978, Mizuno et al., 1992, Sun et al., 1997). Genomic and cDNA sequences have been characterized for rice (Nakamura and Yamanouchi, 1992), maize (Baba et al., 1991; Fisher et al., 1993; Gao et al., 1997) and wheat (Repellin et al., 1997; Nair et al., 1997; Rahman et al., 1997). Sequence alignment reveals a high degree of sequence similarity at both the nucleotide and amino acid levels and allows the grouping into the SBEI, SBEIIa and SBEIIb classes. SBEIIa and SBEIIb generally exhibit around 80% sequence identity to each other, particularly in the central regions of the genes. SBEIIa and SBEIIb may also be distinguished by their expression patterns. SBEIIb is generally specifically expressed in endosperm while SBEIIa is present in every tissue of the plant.
In wheat endosperm, SBEI (Morell et al, 1997) is found exclusively in the soluble fraction, while SBEIIa and SBEIIb are found in both soluble and starch-granule associated fractions (Rahman et al., 1995). In maize and rice, high amylose phenotypes have been shown to result from lesions in the SBEIIb gene, also known as the amylose extender (ae) gene (Boyer and Preiss, 1981, Mizuno et al., 1993; Nishi et al., 2001). In these SBEIIb mutants, endosperm starch grains showed an abnormal morphology, amylose content was significantly elevated, the branch frequency of the residual amylopectin was reduced and the proportion of short chains (<DP17, especially DP8-12) was lower. Moreover, the gelatinisation temperature of the starch was increased. In addition, there was a significant pool of material that was defined as “intermediate” between amylose and amylopectin (Boyer et al., 1980, Takeda, et al., 19936). In contrast, maize plants mutant in the SBEIIa gene due to a mutator (Mu) insertional element and consequently lacking in SBEIIa protein expression were indistinguishable from wild-type plants in the branching of endosperm starch (Blauth et al., 2001), although they were altered in leaf starch. Similarly, rice plants deficient in SBEIIa activity exhibited no significant change in the amylopectin chain profile in endosperm (Nakamura 2002). In both maize and rice, the SBEIIa and SBEIIb genes are not linked in the genome.
In maize, the dull1 mutation causes decreased starch content and increased amylose levels in endosperm, with the extent of the change depended on the genetic background, and increased degree of branching in the remaining amylopectin (Shannon and Garwood, 1984). The gene corresponding to the mutation was identified and isolated by a transposon-tagging strategy using the transposon mutator (Mu) and shown to encode the enzyme designated starch synthase II (SSII) (Gao et al., 1998). The enzyme is now recognized as a member of the SSIII family in cereals (Li et al., 2003). Mutant endosperm had reduced levels of SBEIIa activity associated with the dull1 mutation. No corresponding mutation has been reported in other cereals. It is not known if these findings are relevant to other cereals, for example wheat.
WO94/09144 suggests the use of sense and antisense genes to alter the natural ratios of starch synthase (SS) and SBE in maize. However, no data are presented to substantiate the proposed molecular strategies and there is no suggestion of specifically reducing the activity of SBEIIa.
In potato, down regulation of SBEI alone causes minimal affects on starch structure (Filpse et al., 1996), although further work identified some qualitative changes (Safford et al., 1998). However, in potato the down regulation of SBEII and SBEI in combination increased the relative amylose content much more than the down-regulation of SBEII alone (Schwall et al., 2000).
Two types of debranching enzymes are present in higher plants and are defined on the basis of their substrate specificities, isoamylase type debranching enzymes, and pullulanase type debranching enzymes (Myers et al., 2000). Sugary-1 mutations in maize and rice are associated with deficiency of both debranching enzymes (James et al., 1995, Kubo et al., 1999) however the causal mutation maps to the same location as the isoamylase-type debranching enzyme gene. Representative starch biosynthesis genes that have been cloned from cereals are listed in Table 1.
TABLE 1Starch branching enzyme genes characterized from cereals.SBEType ofSpeciesisoformcloneAccession No.ReferenceMaizeSBEIcDNAU17897Fisher et al., 1995genomicAF072724Kim et al., 1998aSBEIIbcDNAL08065Fisher et al., 1993genomicAF072725Kim et at., 1998SBEIIacDNAU65948Gao et al., 1997WheatSBEIIcDNAY11282Nair et al., 1997SBEIcDNA AJ237897 Bags et al., 1999andSBEI gene)genomicAF002821 (SBEIRahman et al., 1997,pseudogeneAF076680 Rahman et al,, 1999(SBEI gene)AF076679 (SBEI cDNA)SBEIcDNAY12320Repellin et al., 1997SBEIIacDNA AF338432 Rahman et al., 2001and(cDNA)genomicAF338431 (gene)SBEIIbcDNA WO 01/62934andgenomicSBEIIbcDNAWO 00/15810RiceSBEIcDNAD10752Nakamura andYamanouchi, 1992SBEIgenomicD10838Kawasaki et al.,1993RBE3cDNAD16201Mizuno et al., 1993BarleySBEIIa cDNA AF064563 Sun et al., 1998andand(SBEIIb gene)SBEIIbgenomicAF064561 (SBEIIb cDNA)AF064562 (SBEIIa gene)AF064560 (SBEIIa cDNA)
Starch is widely used in the food, paper and chemical industries. The physical structure of starch can have an important impact on the nutritional and handling properties of starch for food or non-food or industrial products. Certain characteristics can be taken as an indication of starch structure including the distribution of amylopectin chain length, the degree and type of crystallinity, and properties such as gelatinisation temperature, viscosity and swelling volume, Changes in amylopectin chain length may be an indicator of altered crystallinity, gelatinisation or retrogradation of the amylopectin.
Starch composition, in particular the form called resistant starch which may be associated with high amylose content, has important implications for bowel health, in particular health of the large bowel. Accordingly, high amylose starches have been developed in certain grains such as maize for use in foods as a means of promoting bowel health. The beneficial effects of resistant starch result from the provision of a nutrient to the large bowel wherein the intestinal microflora are given an energy source which is fermented to form inter alia short chain fatty acids. These short chain fatty acids provide nutrients for the colonocytes, enhance the uptake of certain nutrients across the large bowel and promote physiological activity of the colon. Generally if resistant starches or other dietary fibre is not provided the colon is metabolically relatively inactive.
Whilst chemically or otherwise modified starches can be utilised in foods that provide functionality not normally afforded by unmodified sources, such processing has a tendency to either alter other components of value or carry the perception of being undesirable due to processes involved in modification. Therefore it is preferable to provide sources of constituents that can be used in unmodified form in foods.
More wheat is produced in the world each year than for any other cereal grain crop. Known variation in wheat starch structure is limited relative to the variation available in maize or rice, in part because the transformation efficiency of wheat has lagged behind that for other cereals, and because of the hexaploid nature of breadwheat. The presence of three genomes in Triticum aestivum has a buffering effect by masking mutations in individual genomes, in contrast to the more readily identified mutations in diploid species. Mutants in SBEIIb, corresponding to the amylose-extender phenotypes in maize or rice, have not been characterized in wheat. The phenotype conferred by SBEIIa or SBEIIb mutations in wheat is unknown. Known mutants are for the waxy gene (GBSS, Zhao and Sharp, 1998) and a mutant entirely lacking the SGP-1 protein (Yamamori et al, 2000) which was produced by crossing lines which were lacking the A, B and D genome specific forms of SGP-1 (SSII) protein as assayed by protein electrophoresis. Examination of the SSII null seeds showed that the mutation resulted in alterations in amylopectin structure, deformed starch granules, and an elevated relative amylose content to about 30-37% of the starch, which was an increase of about 8% over the wild-type level (Yamamori et al., 2000). Amylose was measured by colorimetric measurement, amperometric titration (both for iodine binding) and a concanavalin A method. Starch from the SSII null mutant exhibited a decreased gelatinisation temperature compared to starch from an equivalent, non-mutant plant. Starch content was reduced from 60% in the wild-type to below 50% in the SSII-null grain.
WO99/14314 describes the isolation of an SBEIIa gene from Aegilops tauschii, a diploid plant related to wheat, but did not produce wheat with altered starch.
WO 00/15810 describes the cloning of cDNAs for a wheat SBEIIb gene. They did not obtain wheat plants with altered amylose levels and did not teach wheat having starch comprising at least 50% amylose.
WO01/62934 also describes a wheat SBEIIb gene and suggests introducing inhibitors of branching enzyme activity into a wheat plant but does not teach wheat having starch comprising at least 50% amylose.
WO 01/32886 characterized a cDNA encoding a form of SBEI in wheat endosperm. The encoded polypeptide was found to be preferentially associated with A-type starch granules. They did not suppress SBEI activity or show altered starch granule morphology or elevated amylose in wheat.
Therefore, wheat having starch with a proportion of amylose greater than about 50% is unknown. Although high amylose maize and barley varieties are known, products from these cereals have disadvantages compared to a very high amylose wheat for products where wheat is the preferred cereal, for example in bread, pasta or noodles.
Whilst elevated amylose starches of these types are useful, a wheat starch with higher amylose contents is preferred, in particular if associated with improved starch synthesis and other characteristics, for example a reduced need for post-harvest modification. Such starch products are also relatively resistant to digestion and bring a greater health benefit.